Gotta say, more than a bit peeved to open my YouTube subscriptions this afternoon and see Linus showing off the new light switches (which he initially declined to be a part of). Meanwhile, we still haven’t gotten any concrete details about when those that actually supported the campaign are going to get our hands on the units we paid for. I didn’t even know the beta was underway.
I wouldn’t be all that peeved having watched the video. The switch he has doesn’t have the final version mmWave sensor. It’s clearly not ready and I wouldn’t call what he showed beta. It’s alpha at best. It does build brand awareness and demand for Inovelli - which can only improve their profits in the long run and in doing so improve their resources for more (and faster) future development.
I’m curious, is there a reason why not just use a LiDAR instead?
From what I understand from a friend working on some LiDAR-related projects at work, they could potentially work better, without having to go mmWave.
LiDAR is significantly more expensive. Considering the margins on these switches are very low and Inovelli is already at the high end of the price range, it just wouldn’t make financial sense. Also, I believe LiDAR’s waveform is fairly straight and narrow while mmWave is more cone shaped (ie spreads out). The latter is more conducive to a wide angle which is what you’d need to accurately map the presence in a room.
Fair enough, all those points certainly make sense.
Regarding the last one, I’ll ask my friend about it.
Wow - Nice!!! @Eric_Inovelli even had a cameo in it!
Project Update:
We are waiting for tooling to finish so we can test the new switches. In the meantime, I’ve been testing out the sensor still and haven’t had any issues and the zones are working amazing.
Just some clarification around the recent Linus video, it sounds like it was called out, but I want to make it clear that what was shown in the video was the first version of the switch. They were under a time-crunch to complete the video for Homey and they wanted to install our switch for the video.
For those that haven’t seen the video Linus put out, here’s a quick cameo of me walking around the office with the radar tracking me!
As for what you’re seeing, this was just a quick demonstration I setup for Linus’s team. The red rectangle is the detection zone, that I set that encompases my office up until the edge of the desk. I set it at the edge of my desk so you can see me walk outside the zone and presence is no longer detected. The amount of time can be adjusted for when you leave the zone in case you want it instant like in the video or if you want to set it for x minutes after you leave.
The yellow rectangle is more for illustration purposes, but it’s the, “Interference Zone” which you can set to eliminate anything you think will interfere with the sensor (such as fans, windows, etc).
There’s also a, “learning” feature that you can set up where you activate it, turn on all your fans, and then leave the room and the sensor should automatically create interference zones.
There’s an app mentioned in the Linus video. So is that a real thing? How likely will a white series version be?
Love it! Is this done within a hub?
For future devices, I wonder if it’s worth considering a fast config option that would work regardless of hub, and gaster than buttons on the switch.
In the past I’ve played around with building a webapp that could run on any phone & would flash the screen in a pattern that could be detected by a photo sensor on the device, and doing so worked pretty well.
Something like this could be a fallback platforms like matter on apple that don’t offer a better way to configure the device.
Stairway with multiple switches:
I am considering using these for a stairway to a basement with a 9’ ceiling. There is a landing at the top, one in the middle to the outside door, and one at the bottom. Code would require switches in all three places. The switch at the top is on one side, and the others on the other side. My questions are:
- I’m presuming that this scenario would need mmWave sensors on all three switches, due to the narrow stairway and large vertical height difference, yes?
- What are the vertical characteristics of the sensors – what is the vertical beam spread, and are there vertical as well as horizontal coordinates for the zoning?
- In this 4-way scenario (3 switches), could they be wired so that triggering any one sensor would turn on the stairway light (operating directly along the traveler wire, without connection to a hub)?
- Am I correct that the mmWave sensor could be configured to directly control the switch, with no hub attached, or that alternatively, the switch and sensor could be seen as two independent devices with any logic connecting them together or to other devices being controlled by automation at the hub?
Sorry if any of this has already been answered (I did attempt to search othe 650 messages in the topic)!